Again this year, emphasis has been placed on the analysis of electro-physiological data in order to better understand spontaneous fluctuations in the fMRI signal. This work has relied to large extent on primate data obtained by the laboratories of David Leopold (NIMH/NIH) and Naotaka Fujii (BSI, RIKEN, Japan). In a first project, led by Xiao Liu, electro-corticographical data was analyzed to investigate the type of brain activity that occurs in the absence of a task and with the brain at rest. It was found that much of the cortex exhibits synchronized spontaneous events consisting of sequential power changes in distinct frequency bands. These events take place within the timespan of a few seconds and therefore are likely to have a BOLD fMRI correlate, and underlie the signal changes often observed in fMRI-based brain connectivity studies. Comparison of the occurrence of this electrophysiological event during wake, sleep, and anesthesia suggests it reflects momentary drops in brain arousal. Currently, we are investigating the fMRI activity patterns associated with these arousal drops. Preliminary data from primates suggest that these patterns involve much of the cerebral cortex. In a related project, led by Catie Chang, concurrently acquired primate LFP and fMRI data was analyzed to study brain activity changes with changes in behavioral measures of arousal. For this purpose, the position of the eyelid was monitored and correlated with fMRI activity and LFP power. As expected, behavioral arousal changes were reflected in the LFP power, and widespread fMRI activity changes were observed with a prototypical spatial pattern. Ongoing work is geared towards using the fMRI signal as indicator of arousal, and using it to predict fluctuations in behavioral performance. Techniques developed in previous years to quantify brain myelin based on its effect magnetic susceptibility contrast are being applied to study pathological brain changes in Multiple Sclerosis (MS). Xiaozhen Li, in collaboration with the laboratory of Danny Reich, has studied myelin changes in 25 patients with relapsing-remitting MS. Myelin content was inferred from fitting a 3-compartment model to the MRI susceptibility-weighted signal, with one of the compartments representing myelin water and used as a putative marker of myelin content. As expected, myelin water was strongly reduced in MS lesions, suggesting it may be a sensitive marker of myelin loss. Somewhat surprisingly, little difference in myelin water content was found between active (i.e. responding to gadolinium-contrast injection) and in-active lesions, where the latter were thought to represent a relative strong myelin loss.